HYDROLOGICAL PROCESSES
Hydrol. Process. (2008)
Published online in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/hyp.6931
Effect of water table drawdown on peatland dissolved organic
carbon export and dynamics
M. Strack,1 * J. M. Waddington,2 R. A. Bourbonniere2,3 E. L. Buckton,2 K. Shaw,2
P. Whittington4 and J. S. Price4
2
1 Department of Geography, University of Calgary, Calgary, AB, Canada
School of Geography and Earth Sciences, McMaster University, Hamilton, ON, Canada
3 Environment Canada, National Water Research Institute, Burlington, ON, Canada
4 Department of Geography, University of Waterloo, Waterloo, ON, Canada
Abstract:
Peatlands play an important role in the global carbon cycle, and loss of dissolved organic carbon (DOC) has been shown to be
important for peatland carbon budgets. The objective of this study was to determine how net production and export of DOC
from a northern peatland may be affected by disturbance such as drainage and climate change.
The study was conducted at a poor fen containing several pool–ridge complexes: (1) control site–no water table
manipulation; (2) experimental site–monitored for one season in a natural state and then subjected to a water table drawdown
for 3 years; (3) drained site–subjected to a water table drawdown 9 years prior to monitoring. The DOC concentration was
measured in pore water along a microtopographic gradient at each site (hummock, lawn and hollow), in standing water in
pools, and in discharge from the experimental and drained sites. The initial water table drawdown released ¾3 g of carbon
per square metre in the form of DOC, providing a large pulse of DOC to downstream ecosystems. This value, however,
represents only 1–9% of ecosystem respiration at this site. Seasonal losses of DOC following drainage were 8–11 g of carbon
per square metre, representing ¾17% of the total carbon exchange at the experimental study site. Immediately following
water table drawdown, DOC concentrations were elevated in pore water and open water pools. In subsequent seasons, DOC
concentration in the pool declined, but remained higher than the control site even 11 years after water-table drawdown. This
suggests continued elevated net DOC production under lower water table conditions likely related to an increase in vegetation
biomass and larger water table fluctuations at the experimental and drained sites. However, the increase in concentration was
limited to initially wet microforms (lawns and hollows) reflecting differences in vegetation community changes, water table
and soil subsidence along the microtopographic gradient. Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the
Queen in right of Canada.
KEY WORDS
biogeochemistry; carbon cycling; DOC; peatland; hydraulic conductivity; drainage; climate change; wetland
Received 3 January 2007; Accepted 10 October 2007
INTRODUCTION
Northern peatlands play a significant role in the global
carbon cycle and sequester approximately one-third of
the global soil carbon pool (Gorham, 1991) primarily
through peat accumulation in anoxic soils (Clymo, 1984).
Whereas many studies suggest that carbon continues to
accumulate in peat soils (e.g. Waddington and Roulet,
1996; Bubier et al., 1999; Asada and Warner, 2005), most
do not account for hydrologic losses of carbon. Such
losses can be important when determining carbon storage
in peatlands (Billett et al., 2004) and may be increasing
(Freeman et al., 2001; Worrall et al., 2003).
Dissolved organic carbon (DOC) is operationally
defined as organic carbon that is smaller than 0Ð45 µm
in diameter (Thurman, 1985). DOC is found in concentrations ranging from 3 to 400 mg l1 in natural peatlands
and averages 30 mg l1 , whereas DOC export from peatlands is 5–40 g m2 year1 (Thurman, 1985; Moore,
1998). Moreover, the proportion of peatland area within
* Correspondence to: M. Strack, Department of Geography, University
of Calgary, Calgary, AB, Canada. E-mail: [email protected]
a watershed is related to DOC export (Koprivnjak and
Moore, 1992; Mattsson et al., 2005) and DOC loss may
be important for determining peatland carbon balances
(Moore, 1998; Billett et al., 2004). In addition, DOC
plays many important ecological and geochemical roles
both within peatlands and in downstream ecosystems,
affecting acidity, nutrient availability, metal mobility and
light penetration in aquatic habitats (Steinberg, 2003).
DOC export from peatlands occurs in two stages:
(1) the production of DOC and (2) export. High net
production in peatlands is due to the largely anaerobic
soil conditions (e.g. Moore and Dalva, 2001). Increases in
primary productivity can result in higher concentrations
of peatland DOC (Freeman et al., 2004). DOC export is
controlled by site hydrology (Hinton et al., 1997; Fraser
et al., 2001; Pastor et al., 2003) and is generally greater
from peatlands with higher measured discharge (Fraser
et al., 2001; Freeman et al., 2001). Storm events also play
a large role in DOC export, causing flushing of DOC-rich
water from peatlands to downstream ecosystems (Hinton
et al., 1997). Thus, any disturbance altering factors
affecting DOC production or hydrology could potentially
change the quantity and chemistry of exported DOC.
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
M. STRACK ET AL.
Northern peatland water tables may be lowered by
drainage (e.g. Moore and Roulet, 1993; Price, 2003) or
climate change (Roulet et al., 1992). Many peatlands
are drained for forestry, agriculture or peat extraction
(Moore, 2002; Joosten and Clarke, 2002). This process
involves the construction of ditches, thereby increasing
both peatland discharge (Van Seters and Price, 2002) and
DOC export. Also, lowered water tables will initially
cause surface subsidence as a result of peat compression
(Price and Schlotzhauer, 1999; Price, 2003). This, in turn,
decreases hydraulic conductivity and specific yield and
increases the magnitude of water table fluctuations (Price
and Schlotzhauer, 1999; Price, 2003; Whittington and
Price, 2006). Through time, continued subsidence occurs
as a result of peat oxidation (Schothorst, 1977; Price and
Schlotzhauer, 1999; Price, 2003).
Under climate change, persistent water table drawdown
may result in the maintenance of the water table at a
level, relative to the surface, that is similar to pre-drainage
conditions as a result of continued peat subsidence (e.g.
Whittington and Price, 2006). In addition, a change in the
vegetation community may occur as observed in naturally
(Foster et al., 1988) and artificially drained peatlands
(Minkkinen et al., 2002; Strack et al., 2006). If this
results in increased vegetation productivity then it may
enhance production and export of DOC (Freeman et al.,
2004). Thus, long-term DOC dynamics are influenced by
ecological as well as physical and hydrological changes.
This complex interplay of processes leads to questions
regarding the applicability of short-term drought studies
when considering climate change scenarios.
Therefore, the aim of this study was to compare the
concentration of DOC in pore water and surface water,
and DOC export between a natural peatland, a peatland
with newly imposed drainage and water table drawdown,
and a peatland with the water table lowered 9 years prior
to the study. Specifically, the objectives were (1) to determine the effect of peatland drainage on DOC export
during the initial ditching, subsequent two growing seasons and following 9 to 11 seasons of persistent water
table drawdown, (2) to determine the impact through
time of persistent water table drawdown on DOC dynamics in peatland pore water and open water pools, (3) to
determine whether changes in pore water DOC dynamics
induced by water table drawdown varied between peatland microforms, and (4) to investigate ecohydrological
controls, such as water table and vegetation productivity, on peatland DOC dynamics. The results will provide
insight regarding the interaction between hydrology and
ecology for controlling peatland DOC dynamics, provide
information about the potential response of peatland DOC
dynamics to climate change, and provide an estimate of
DOC exported via peatland drainage.
MATERIALS AND METHODS
Study area
The study site is a poor fen fragment (see Kellner et al. (2004)) with peat depth of 0Ð8–1Ð5 m underlain by clay mineral soil located in St-Charles-deBellechasse, Québec (46° 400 N, 71° 100 W). The 30-year
normal (1971–2000) precipitation and temperature from
May to September for the region are 590 mm and
15Ð5 ° C respectively (climate data available from Environment Canada at http://climate.weatheroffice.ec.gc.ca).
The study was carried out between 2001 and 2004, during which time 2001 and 2002 were drier than normal
and 2003 and 2004 were close to the long-term average
(Table I).
To examine the effects of peatland drainage on DOC
dynamics through time, three pool–ridge complexes were
compared (Figure 1). The first did not undergo any water
table manipulation (control, C). The second (experimental, E) pool–ridge complex was monitored for one season
(2001) and then subjected to a water table drawdown with
a ditch connecting the pool to the larger drainage network. The third pool–ridge complex (drained, D) underwent water table drawdown by the peatland operators
in 1993. Sampling was conducted along the microtopographic gradient at each site, with the driest ridge locations referred to as hummocks, intermediate locations as
lawns, low-lying areas as hollows and open water areas as
pools. At D, the water table drawdown almost completely
eliminated the open water area; flooding only occurred in
the spring and after intense summer storms. Also, water
table drawdown has resulted in observable changes to
the vegetation community at E and D. In general, hummocks experienced a reduction in moss cover, lawns
Table I. Climatic data from Québec City (30 km from the St Charles-de-Bellechasse research site) during the four seasons of the
study and the 30-year normal (1971–2000) for the same location. Precipitation gives monthly totals and temperature, monthly mean
2001
2002
2003
2004
Normal
Precipitation (mm)
Temperature (° C)
Precipitation (mm)
Temperature (° C)
Precipitation (mm)
Temperature (° C)
Precipitation (mm)
Temperature (° C)
Precipitation (mm)
Temperature (° C)
May
June
July
August
September
Seasonal
58Ð4
13Ð5
107Ð6
9Ð5
100Ð6
10Ð8
122Ð9
10Ð5
105Ð5
11Ð2
103Ð1
17Ð6
67Ð7
14Ð9
97Ð4
17Ð1
117Ð0
14Ð9
114Ð2
16Ð5
72Ð2
17Ð7
63Ð4
19Ð8
129Ð6
18Ð4
168Ð8
19Ð5
127Ð8
19Ð2
105Ð7
18Ð8
11Ð8
19Ð4
153Ð2
18Ð1
71Ð6
17Ð7
116Ð7
17Ð6
93Ð8
14Ð0
107Ð7
15Ð6
85Ð0
15Ð3
115Ð6
13Ð7
125Ð5
12Ð5
433Ð2
16Ð3
358Ð2
15Ð8
565Ð8
15Ð9
595Ð9
15Ð3
589Ð7
15Ð5
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
WATER TABLE DRAWDOWN AND PEATLAND DOC DYNAMICS
experienced an increase in vascular plant cover, and hollows/pools experienced an increase in moss and vascular
plant cover relative to C (Table II). For more information
on vegetation changes, see Strack et al. (2006), Strack
and Waddington (2007) and St-Arnaud (2007).
Across the site there is a gradient of approximately
50 cm per 100 m, with C at the highest elevation
followed by E and then D. Hydrology and DOC dynamics
were monitored at all sites between July 2001 and August
2004, with measurements generally carried out between
May and September of each study season.
Dissolved organic carbon concentration
DOC concentration was determined in surface and
pore water at each site and in discharge from E and
D. Pore water samples were collected from piezometer
nests installed throughout the peat depth and along the
microtopographic gradient at each site. One nest of
piezometers was placed at a hummock, lawn and hollow
microform at each site. Samples were collected monthly
N
Experimental
E
x
∗
xx
Control
C
x xx
xxx
xxx
Drained
D
x Piezometer nest hydraulic head and conductivity
Piezometer nest pore water DOC
0
10 m
* Meteorological station
Figure 1. St Charles-de-Bellechasse experimental fen showing the extent
of the hollow area at each of the C, E and D sites. Piezometers used
for pore water DOC sampling and determination of subsurface flow are
identified. The unlabelled hollow is an additional natural pool–ridge area
used to assess influx of water into and between natural zones
Table II. Average vegetation cover at each site along the microtopographic gradient in 2004 following 3 and 11 seasons of water
table drawdown at the experimental and drained sites respectively
Site
Control (C)
n
7
19
29
Experimental (E) 13
18
21
Drained (D)
16
12
23
Microform
Hummock
Lawn
Hollow/Pool
Hummock
Lawn
Hollow/Pool
Hummock
Lawn
Hollow/Pool
Cover, average š SD (%)
Vascular
Moss
51 š 21
25 š 20
13 š 8
65 š 26
48 š 31
53 š 31
60 š 27
75 š 23
51 š 24
83 š 30
74 š 38
7 š 18
59 š 42
61 š 38
6 š 19
43 š 36
62 š 35
71 š 29
throughout the study period; however, in 2001, only
lawn areas were sampled and no pore water samples
were collected from D. All sites had piezometers with
20 cm slotted intakes centred at depths 25, 50 and 75 cm
below the surface with 100 cm also present at E, and
100 and 125 cm also present at C due to greater peat
depths. Below the piezometer intake was a reservoir
with a volume of at least 100 ml to collect water for
determination of DOC concentration. Prior to sampling,
the reservoir was pumped dry and allowed to refill.
Fresh pore water was then collected in a clean Nalgene
bottle after first rinsing the bottle with a small amount
of sample. Some piezometers refilled quickly and could
be sampled within 1 day of pumping, whereas others
refilled more slowly (particularly at depth) due to the
low hydraulic conductivity of the peat. If at least 50 ml
of sample could not be collected after 3 days, then the
DOC concentration was not determined for that location.
Surface water from the pools and drainage outflow
pipes was sampled every 2 to 3 weeks over the 2002
and 2003 field seasons and monthly in 2001 and 2004.
Nalgene bottles were rinsed with pool or discharge water
and then a sample was collected.
All samples were filtered first with a 1Ð5 µm glass-fibre
filter (Whatman 934-AH) followed by a 0Ð7 µm glassfibre filter (Whatman GF/F) and were then refrigerated
and transported to the National Water Research Institute
where they underwent chemical analysis. The definition
of dissolved is most often made using 0Ð45 µm poresize membrane filters, but polymeric filters often bleed
DOC into samples. The use of pre-rinsed glass-fibre filters eliminates bleed, but compromises the definition of
dissolved somewhat because these filters are characterized as having nominal rather than exact pore sizes. The
GF/F filters (Whatman) used in this study were also used
by others in the field (e.g. Qualls and Haines, 1991,
Mattsson et al. 1998, Brooks et al., 1999) and have a
nominal pore size of 0Ð7 µm. Filtered samples from this
study held for archival purposes at 4 ° C remained stable
showing no flocculation of colloids during storage for up
to 3 years.
DOC concentrations were determined using an Apollo
9000 Total Carbon Analyzer (Techmar-Dohrmann)
employing platinum on aluminium oxide catalyst at
900 ° C. Samples were routinely diluted using EPure
(Barnstead) water to ensure they fell within the calibration range (carbon concentration [C] D 0–20 mg l1 ).
Samples were acidified with 20% phosphoric acid just
prior to analysis and were then sparged for 5 min with O2
carrier gas to remove dissolved inorganic carbon. Each
sample was injected six times and the four nearest replicates were averaged by the Apollo software. Within each
analysis run of 70 vials, a calibration set, five blank and
five check standards were analysed along with the samples. Blank and drift correction curves were constructed
by regressing the deviations from the true values of blank
and check standards respectively against vial sequence
number, and these curves were applied to the initial
results on each sample, which were then finally corrected
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
M. STRACK ET AL.
for dilution. Standard deviations for replicate DOC measurements fell between [C] D 0Ð03 and 1Ð46 mg l1 with
an average standard deviation of [C] D 0Ð4 mg l1 .
Inter-microform dissolved organic carbon flux
In order to determine the flux of DOC between sites
and from hummock to hollow microforms within each
site, water flow was computed according to Darcy’s law
and multiplied by pore water DOC concentrations:
F D [DOC]KA dh/dl
1
where [DOC] is the average concentration of DOC
in 2004 for pore water at 50 cm depth, K is the
geometric mean of hydraulic conductivity throughout
the peat profile, A is the cross-sectional area of flow,
and dh/dl is the hydraulic gradient from hummock to
hollow. One additional transect of piezometer nests was
installed in 2004 from hummock to pool at the upslope
and downslope edge of each site. Hydraulic conductivity
was determined on several occasions throughout 2004
at lawns at 50, 75 and 100 cm depths using bail tests
(Freeze and Cherry, 1979). The hydraulic gradient was
determined as the difference in absolute water table
position between the hummock and hollow across the
horizontal distance of the piezometer transect. The crosssectional area of flow was estimated based on the
perimeter of the lawn zone at each site multiplied by
an average peat depth of 1Ð4 m, 1Ð2 m and 0Ð8 m at
the C, E and D respectively. Uncertainty in this flux
was determined according to Bubier et al. (1999) using
a first-order Taylor series and assuming that uncertainty
in the parameters was independent. Uncertainties in
K and [DOC] were based on the standard deviation
of measurements taken throughout the study season,
whereas the uncertainties for A, dh and dl were estimated
as 10%, 0Ð2 cm and 0Ð5 m respectively.
Since water generally flowed from hummock to hollow
on the upslope edge of each site and hollow to hummock
on the downslope edge, the net flux of DOC was
determined as the balance of these and expressed on an
area basis relative to the estimated area of the hummocks
and hollow/pool at each site. Uncertainty in the net flux
was determined as the square root of the sum of the
variances of the DOC input and output.
Hydrology and dissolved organic carbon export
Water table position was monitored manually during
all study seasons in 2Ð5–3 cm i.d. PVC wells installed
systematically throughout the study sites. In 2001, water
table was measured continuously with a counterbalanced
pulley on a potentiometer at a central meteorological
station. This was connected to a CR21X data logger
(Campbell Scientific, Alberta, Canada) measured each
minute and averaged every 20 min. In 2002–2004, the
water table was measured continuously at lawns at each
site with counterbalanced pulleys and potentiometers
connected to CR10X data loggers (Campbell Scientific).
RDS (Remote Data System) wells were also used at each
site to measure water table position in 2004.
Discharge from the experimental and drained sites
were funnelled into horizontal PVC pipes which provided
outflow to the drainage ditches. Discharge was measured
several times a week from 2002 to 2004 at these
outflow pipes for the experimental and drained sites.
Flow was collected into a graduated container over a
timed interval and discharge was calculated. The reported
values represent averages of three measurements. DOC
export was determined for E in 2002–2004 and for
D in 2003 and 2004. To determine export, manual
discharge measurements were regressed against water
table position and this relationship was used to estimate
daily discharge. DOC concentrations were interpolated
between measurement dates and multiplied by calculated
discharge rates to determine daily export values (modified
from Hinton et al. (1997)). Owing to limited discharge
data for D in 2002, export was not determined for this
season. Since export was determined as the product of
discharge and DOC concentration in the discharge water,
uncertainty was approximated as (Devito et al., 1989)
2
2
2
2
SE2 D Q2 SDOC
C [DOC]2 SQ
C SDOC
SQ
2
where SE is the standard deviation of the export estimate,
SQ is the standard error of the water table–discharge relationship, SDOC is the standard deviation of discharge DOC
concentrations, [DOC] is the average DOC concentration
in the discharge and Q is the average daily discharge.
The catchment area for each pool (C, E and D) was
determined based on peat surface topography. Since surface water discharge was of interest, catchment area was
determined based on surface topography, as it has been
shown that, within patterned peatlands, little surface flow
occurs when surface pools are disconnected (e.g. Quinton and Roulet, 1998), which was the case throughout
study period. Although some subsurface water movement
occurs (as determined above) and may contribute to surface water outflow, its magnitude is generally less than
10% of surface discharge. Also, since surface discharge
ceases soon after standing water disappears, subsurface
flow likely plays a minimal role in surface water loss. The
estimated catchment size for each of the sites is ¾900 m2 .
The total export value (grams of DOC-C) for the season
was divided by the catchment area size to obtain export
on an area basis.
Since discharge was not determined during the initial
drainage at E in 2002, the mass of DOC lost per unit
area was determined by multiplying the depth of water
lost (¾20 cm) by peat porosity (¾0Ð9 according to Baird
and Gaffney (1995)) and average DOC concentration
measured in the drainage water.
Statistical analysis
Differences in water table position between sites and
DOC concentrations between sites and microforms were
assessed with one-way analysis of variance (ANOVA)
with Tukey pairwise comparisons at a family error rate
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
WATER TABLE DRAWDOWN AND PEATLAND DOC DYNAMICS
of 0Ð05. Regression lines between DOC concentrations
and environmental variables were fitted using the leastsquares method and were considered significant when
p < 0Ð05. All analyses were completed in Minitab 14Ð1
(Minitab Inc.). Because all piezometers could not be sampled on each sampling date due to low recharge rates at
some locations, sample size for statistical analysis varied
between microforms and sites. In all cases, microform
averages are based on greater than six samples, and in
general the sample size was greater than 12.
RESULTS
Water table position
Based on manual measurements at lawns in 2001 (prior
to water table manipulation at E), average plus/minus
standard deviation water table position was significantly
deeper (ANOVA, p < 0Ð05) at C (9Ð5 š 3Ð2 cm, n D
23, where a negative value indicates below the soil surface) than the E (5Ð2 š 2Ð5 cm, n D 23). D had a water
table of 14Ð0 š 1Ð9 cm (n D 17) that was significantly
lower than both C and E. Following water table drawdown on day 161 of 2002, the water table dropped rapidly
at E and remained significantly deeper than C throughout the remainder of the study (Figure 2). Fluctuations in
water table position were related to precipitation events;
however, the range of these fluctuations varied between
sites. In 2004, the range for water table position during
the growing season was 11Ð6 cm, 14Ð1 cm and 16Ð9 cm at
C, E and D respectively (see also Whittington and Price
(2006)).
Pore water dissolved organic carbon concentrations
Average plus/minus standard deviation pore water concentrations were 18Ð7 š 3Ð8 mg l1 , 24Ð4 š 9Ð5 mg l1
and 28Ð1 š 12Ð0 mg l1 at C, E and D respectively (n D
61, 69, 41 at C, E and D respectively) during the first
year of the study. Thus, pore water concentration during this time was significantly lower at C than at E,
which was lower than at D (ANOVA, p < 0Ð05). Because
DOC concentrations were initially significantly higher at
E than at C, it is difficult to evaluate changes in pore
water concentration related to water table drawdown statistically. However, in all seasons following drainage,
average E pore water concentrations were elevated above
their initial levels and similar to those at the drained site
(Figure 3).
In a natural state (C and E prior to water table drawdown), pore water DOC concentration varied significantly along the microtopographic gradient, this being
highest at hummocks, intermediate at lawns and lowest
at hollows (ANOVA, p < 0Ð05). There was no consistent pattern of DOC concentrations with depth at any
microform; however, DOC concentration increased with
depth at E and D hollows in 2002. This pattern was much
less apparent in 2004 (Figure 4). Following water table
drawdown (E after water-table drawdown and D), there
was no consistent pattern for DOC concentration between
microforms.
At E there were few significant changes in DOC
concentration before and after water table drawdown.
The only statistically significant changes were at 50 cm
at lawns and hollows, which had higher concentrations
in the first season after water-table drawdown compared
with initially. In general, there was a trend for pore
water DOC concentrations to decline at the hummock
and increase at lawn and hollow following water table
drawdown (Figures 3 and 4).
Inter-microform dissolved organic carbon flux
Geometric mean hydraulic conductivity in the peat
profile declined from approximately 104 cm s1 at C
to 104 –105 cm s1 at E and 105 –106 cm s1 at D.
Water table drawdown resulted in subsidence of the peat
surface, with a greater change in peat volume occurring
in low-lying areas than in neighbouring hummocks. This
uneven subsidence resulted in steeper hydraulic gradients
from the hummock to the pool at the E and D. The
combined effect of these changes on site properties
resulted in decreased fluxes of water between hummocks
and pools from about 2 m3 day1 at C to 0Ð3 m3 day1
at E and 0Ð06 m3 day1 at D. The change in total fluxes
of DOC between microforms followed the hydrological
changes.
Despite the fact that more DOC moved between
microforms at C, the inputs and outputs of DOC were
nearly equivalent, whereas more water and DOC entered
the pool at E than left from the downslope edge, resulting
in a source of DOC-C to the pool of 5 g m2 over the
study season (Table III). Owing to the low hydraulic
conductivity at D, very little carbon moved between
microforms.
As a result of the overall gradient of the peatland,
approximately 4 g m2 DOC-C were transferred from the
C to E, whereas ¾0Ð3 g m2 DOC-C moved between E
and D during the 2004 study season. Comparing the flux
of carbon between C and a neighbouring natural pool,
0Ð4 g m2 DOC-C moved between natural pools over
the same period.
Surface water dissolved organic carbon concentrations
The prior water table drawdown DOC concentration of surface water in pools was not significantly
different between C and E, with average plus/minus
standard deviation values of 16Ð2 š 4Ð71 mg l1 (n D
12) and 22Ð4 š 6Ð1 mg l1 (n D 11) respectively. During this period, average DOC concentration at the D
was 25Ð5 š 3Ð4 mg l1 (n D 5), which was significantly
higher (ANOVA, p < 0Ð05) than C but not E (Figure 2).
During the first season after water-table drawdown, E
pool concentrations increased substantially, averaging
40Ð5 š 22Ð0 mg l1 (n D 8), whereas C and D pool concentrations were on average 19Ð0 š 2Ð1 mg l1 (n D 10)
and 30Ð0 š 6Ð1 mg l1 (n D 5) respectively. During this
time, DOC concentration at E pool was significantly
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
M. STRACK ET AL.
Figure 2. Water table position at lawns (solid symbols) and pool DOC concentrations (open symbols) throughout the study period. The vertical
dashed line indicates the date of drainage at the E pool. The dotted line gives manual water table measurements during periods when continuous
measurements were unavailable. Pool DOC concentrations in 2003 on day 121 were 12Ð0 mg l1 , 19Ð4 mg l1 and 18Ð0 mg l1 at C, E and D
respectively (not shown)
higher than C and D. Large variability observed in E
pool DOC concentrations appear to be related to dilution by precipitation (Figure 2). By the second season, post-water-table drawdown (2003) E concentrations declined slightly but remained significantly higher
than C. The average concentrations at C, E and D
were 12Ð6 š 1Ð9 mg l1 , 27Ð2 š 6Ð6 mg l1 and 24Ð2 š
8Ð3 mg l1 (n D 10) respectively. During the third season
after drawdown (2004) there were no significant differences between pools.
Discharge and dissolved organic carbon export
Throughout the 2 days following drainage at E, DOC
concentration in discharge water was determined every
few hours. The concentration increased from 17Ð3 to
21Ð7 mg l1 over the first 3 h after drainage and declined
to 20Ð7 mg l1 over the following 2 days. Based on
the water table decline, pool catchment area and peat
porosity, it was estimated that 1Ð62 ð 105 l (180 mm)
of water was lost during this initial drainage. Assuming
that this water had an average concentration equivalent
to the mean of all measurements taken during the first
week following the initial ditch opening, i.e. 20Ð3 mg
l1 , an estimated DOC-C of 3Ð7 š 1Ð7 g m2 was lost
during this period from E. During the remainder of the
growing season (days 166–230), DOC-C loss from the E
was 5Ð2 š 1Ð1 mg m2 (Table IV). Despite the fact that
2004 had the greatest amount of precipitation during the
study (Table I), discharge was highest in 2003 (Table IV).
This was primarily due to precipitation distribution. One
large storm in August resulted in a large proportion of the
discharge in 2003, whereas precipitation and discharge
in 2004 were more evenly distributed. During these two
growing seasons (days 128–230), DOC-C export from
the E and D sites were similar, ranging between 8Ð4 and
11Ð3 g m2 (Table IV).
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
WATER TABLE DRAWDOWN AND PEATLAND DOC DYNAMICS
Figure 3. Seasonal average pore water DOC concentration at (a) hummocks, (b) lawns, (c) hollows and (d) average site conditions for C, E and D.
Error bars represent one standard deviation. Sample size n varied between microforms and sites as not all sites could be measured on each sampling
date due to low water tables or slow recharge rates. All means are based on greater than five values and ‘total’ values are the mean of greater than
30 samples at all sites
DISCUSSION
Effect of water table drawdown on dissolved organic
carbon dynamics
Drainage and persistent water table drawdown resulted
in higher DOC concentrations in surface water and pore
water, and increased DOC export. Evidence from E
reveals that these changes take place during the first season following water table drawdown, whereas D suggests
that they persist for many seasons. Drainage increased
DOC export due to an increase in growing season surface discharge from the peatland from near zero to a
range of 2–6 mm day1 . In addition, DOC export also
increased due to increased DOC concentration in surface
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
M. STRACK ET AL.
Figure 4. Profile of seasonal average pore water DOC concentrations before water table drawdown (2001 and May 2002) and during the third season
after water table drawdown (2004) at (a) hummocks, (b) lawns and (c) hollows
and pore water. Our estimate of DOC export in surface water was limited to the growing season and we
have missed export of DOC resulting from snowmelt and
autumn discharge. Since snowpack depth should be similar across all sites and spring and autumn pool DOC
concentrations are similar between C, E and D (Figure 2),
then it is likely that DOC lost during this non-growing
season period is similar among sites. Because we were
interested in differences between C, E and D, the growing
season was most important for understanding these differences; however, year-round monitoring of peatland DOC
export would be desirable. Despite the fact that growingseason loss of DOC-C via surface water is within the
range (5–40 g m2 year1 ) reported for natural peatlands (Moore, 1998), it is clear that drainage has greatly
increased DOC export at E and D relative to C.
Initially high pool DOC concentrations at E following drainage were likely caused by disturbance,
which has been observed to enhance net DOC production (Lundquist et al., 1999; Blodau and Moore, 2003).
Increased concentrations may also result from changing
the source of water to the pool. Following water table
drawdown, uneven peat subsidence resulted in steeper
hydraulic gradients, potentially increasing the input of
higher DOC peat pore water to the pool.
During the second season after water table drawdown,
DOC concentration in the E pool declined, but they
remained similar to D and elevated above C. Elevated
DOC concentrations were also observed in pore water
following drainage, and this was maintained throughout the study period. Low concentrations in 2004 may
be linked to the larger amount of precipitation (Table I)
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
WATER TABLE DRAWDOWN AND PEATLAND DOC DYNAMICS
Table III. Hydrologic parameters and inter-microform transfer of water and DOC at control (C), experimental (E) and drained
(D) sites in 2004. Negative gradients dh/dl indicate a slope from the hollow to the hummock and negative values of DOC flux
indicate a loss of carbon from the microform. Hydraulic conductivity K values are the geometric means of K throughout the peat
profile. Hydraulic gradients are the average of measurements made twice weekly over the growing season
Location
Control
Upslope
Downslope
Total
Experimental
Upslope
Downslope
Total
Drained
Upslope
Downslope
K (cm s1 )
3 ð 104
nD8
6 ð 103
nD5
4 ð 105
n D 10
2 ð 104
nD2
1 ð 105
n D 10
3 ð 106
n D 10
dh/dl
Subsurface discharge (l day1 )
2 ð 102
n D 21
1 ð 102
n D 21
8 ð 102
n D 21
6 ð 103
n D 21
7 ð 103
n D 21
6 ð 102
n D 21
DOC flux (mg m2 day1 )
Hummock
Hollow
2 ð 103
60 š 200
20 š 90
9 ð 102
30 š 100
10 š 50
30 š 200
10 š 100
4 ð 102
70 š 500
80 š 200
1 ð 102
30 š 2000
30 š 700
40 š 2000
50 š 700
1 ð 101
4 š 1
6š1
3 ð 101
2š4
2 š 2
2š4
4š2
Total
Table IV. Calculated DOC export in surface water from 2002 to 2004 at experimental and drained sites. Surface water discharge
was never observed at the control (C) site during the growing season; therefore, both discharge and export are assumed to be zero.
Limited discharge measurements at the drained site in 2002 preclude export determination during that season. Values are given as
average plus/minus standard error. Standard error for discharge and export are computed as described in the text
Draineda (D)
Experimental (E)
Discharge
(mm)
Initial drainage (days
161–166)
Following drainage
(days 167–230)
2002 growing season
total (days
161–230)
2003 growing season
(days 128–230)
2004 growing season
(days 128–230)
a n.m.:
180 š 70
236 š 271
416 š 341
646 š 700
447 š 354
Average
discharge
[DOC] (mg l1 )
DOC-C
export
(mg m2 day1 )
Discharge
(mm)
Average
discharge
[DOC] (mg l1 )
20Ð3 š 0Ð4
n D 13
26Ð6 š 1Ð4
n D 15
23Ð9 š 1Ð0
n D 28
617 š 283
n.m.
n.c.
n.c.
81 š 116
n.m.
n.c.
n.c.
127 š 130
n.m.
30Ð2
n.c.
13Ð8 š 1Ð8
nD7
10Ð7 š 0Ð7
nD4
110 š 105
342 š 504
91 š 38
330 š 449
22Ð4 š 1Ð8
nD7
14Ð1 š 1Ð4
nD4
DOC-C
export
(mg m2 day1 )
104 š 117
82 š 63
not measured; n.c.: not calculated.
resulting in dilution and flushing of DOC in surface
water (Schiff et al., 1998; Moore and Dalva, 2001;
Tóth, 2002). In fact, according to Whittington (2005),
seasonal values of precipitation minus evapotranspiration were 10 mm, 7 mm and 72 mm in 2002, 2003
and 2004 respectively; these are inversely correlated
with average surface water DOC concentrations at C
and D.
Despite the fact that higher DOC concentrations at E
may be partially due to an increase in the inter-site and
inter-microform transfer of water and DOC, evidence
from D suggests that this enhanced subsurface flow is
not persistent. Over time, subsurface flow is reduced by
continued reductions in hydraulic conductivity caused
by ongoing peat subsidence related to shrinkage and
peat oxidation (Schothorst, 1977; Price, 2003). Moreover,
enhanced hydraulic gradients may be minimized over
time as peat is lost to soil respiration at hummocks
and accumulated in hollows and pools via enhanced
vegetation productivity (Strack et al., 2006). Finally,
the uncertainty in the absolute value of net subsurface
DOC exchange is high, particularly at E, making it
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
M. STRACK ET AL.
difficult to assert that the flow of dissolved carbon
to hollows has been increased in the short term by
water table drawdown. Therefore, maintenance of higher
concentrations of DOC in pore water and surface water
at E and D suggests that water table drawdown results in
persistent elevated rates of net DOC production.
Controls on net dissolved organic carbon production
Shifts in the concentration of DOC result from changes
in the balance between DOC production and consumption. Because we determined only concentration, we are
unable to assess whether concentration changes result
specifically from changes in production or consumption,
but know only that net production of DOC has increased.
This net production of DOC in organic soils may be controlled by temperature, soil moisture, soil solution chemistry, vegetation community and site hydrology (Moore
et al., 1998; Kalbitz et al., 2000); however, the relative importance of these controls is not clear (Kalbitz
et al., 2000). Although several field studies have observed
higher DOC concentrations in relation to higher temperatures (Moore, 1987; Bourbonniere, 1989; Moore and
Dalva, 2001; Waddington and Roulet, 1997), water table
drawdown had little effect on thermal regime in this study
(see Strack et al. (2006)), suggesting that temperature is
not the major factor influencing DOC concentrations following water table drawdown.
Water table position can also affect net production
of DOC, as anaerobic conditions lead to less efficient
decomposition of organic matter, increasing the production of water-soluble intermediate metabolites (Mulholland et al., 1990; Kalbitz et al., 1997). Water table drawdown exposes the upper peat profile to aerobic conditions,
favouring CO2 over DOC as a metabolic end product
(Freeman et al., 2004). However, following drainage,
peat subsidence helped to maintain the water table closer
to the surface than would be observed in a rigid soil (e.g.
Whittington and Price, 2006). There is spatial variability
in the response of DOC dynamics to water table drawdown because of variability in initial water table position and amount of peat subsidence between microforms.
Following water table drawdown, DOC concentrations
increased only at relatively wet microforms (lawns and
hollows). Hummocks have experienced less peat subsidence, more water table drawdown (Whittington and
Price, 2006) and reveal a trend of reduced pore water
DOC concentrations (Figures 3 and 4). Since the water
table at E and D hummocks is generally 30 to 40 cm
below the surface, a large aerobic zone in the peat profile
may result in the dominance of CO2 as the end product of
decomposition. This is supported by increases in ecosystem respiration observed at D hummocks (Strack et al.,
2006). The enrichment of DOC in lawn and hollow pore
water is likely linked to larger water table fluctuations and
a changing vegetation community (discussed below).
Aside from the importance of the water table position, water table fluctuation can also play an important
role in DOC production. In a review of literature, Kalbitz et al. (2000) assert that rewetting following a dry
period results in increased DOC concentrations in both
field and laboratory studies. Microbial utilization of DOC
is reduced during dry periods, and turnover of microbial biomass is enhanced upon rewetting, resulting in
enhanced DOC release (Lundquist et al., 1999). Also,
several studies have observed that DOC production in wet
soils decreases with time (Christ and David, 1996; Blodau and Moore, 2003) potentially resulting from a build
up of DOC, which limits its continued production through
the inhibition of anaerobic decomposition or DOC dissolution from an adsorbed phase. Water table fluctuations
can flush DOC from stagnant soil horizons and may promote further DOC production. In the current study, the
seasonal range of water table position increased with time
after water-table drawdown (i.e. D has larger water table
fluctuations than E). During periods of drought, water
table is drawn down further at E and D than at C, and during subsequent precipitation events the water table rises
more rapidly at the former sites (Figure 2). This different hydrological response results from decreased porosity
caused by soil subsidence (Whittington and Price, 2006).
These rapid wetting events through a larger area of unsaturated peat have likely increased net DOC production at
E and D.
Additionally, increased vegetation productivity can
increase net DOC production (Lu et al., 2000; Kang
et al., 2001). In the current study, water table drawdown
resulted in ecological succession at both E and D (StArnaud, 2007; Strack and Waddington, 2007). At E
and D, the Sphagnum cover at hummocks was reduced,
the sedge cover at lawns increased and the hollows
were colonized by Sphagnum (Table II). Thus, vegetation
productivity at hummocks was reduced, whereas lawns
and hollows have become more productive at E and
D (Strack et al., 2006). These changes in vegetation
productivity are mirrored by the trends in pore water
DOC concentrations. This suggests that enhanced inputs
of fresh organic matter by the changing vegetation
community are partially responsible for the higher DOC
concentrations observed across the experimental and
drained sites following water table drawdown.
Dissolved organic carbon in the context of the carbon
cycle
It is well known that the carbon balance of peatlands
is often very close to neutral (Bubier et al., 1999),
with many sites storing carbon in some seasons and
releasing carbon during others, often in relation to
climatic conditions (e.g. Shurpali et al., 1995; Joiner
et al., 1999). Thus, relatively small losses of carbon
as DOC may be important in relation to the overall
peatland carbon budget. For 2003 and 2004 at E, DOC
export accounted for approximately 17% of the total
carbon budget. Therefore, as has been found in other
studies (Waddington and Roulet, 2000; Billett et al.,
2004), losses of carbon as DOC should be considered
when determining net peatland carbon balance.
Moreover, the rate of carbon cycling can vary greatly
between microforms within a peatland due to varying
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
WATER TABLE DRAWDOWN AND PEATLAND DOC DYNAMICS
environmental conditions. Several studies suggest that
carbon accumulation is highest at hummocks, whereas
hollows or pools are generally sources of carbon to the
atmosphere (Moore, 1989; Belyea and Clymo, 2001).
However, these studies generally ignore hydrologic fluxes
of carbon, particularly between microforms. We have
found that hummocks tend to lose carbon as DOC while
pools gain DOC, and this may partially account for
differences in carbon accumulation rates determined from
soil–atmosphere gas fluxes alone.
Implications of drainage and climate change on
peatland dissolved organic carbon dynamics
The initial drainage of the 900 m2 E site released
3Ð3 š 1Ð5 kg DOC over several days. In general, peatland
drainage for peat extraction involves deeper drainage over
much larger areas. According to Cleary et al. (2005),
¾760 ha of peatland in Canada entered into production
for horticultural peat each year between 1990 and 2003.
Considering the standard method of 30 m ditch spacing,
the water table is lowered from its initial position by
an average of 50 cm (Boelter, 1972; Hillman, 1992).
Assuming a porosity of 0Ð9 and a DOC concentration
(as observed in this study) of ¾20 mg l1 , this drainage
results in an increase in DOC export of 7 ð 107 g of
carbon per year, or ¾9 g of DOC per square metre
per year in the affected catchments. This represents a
substantial load to downstream ecosystems, delivered as
a pulse, with the potential to alter their physical and
chemical characteristics such as acidity, light penetration,
and metal and nutrient availability (Steinberg, 2003).
Although the percentage of peatlands affected by active
drainage is small in Canada, the majority of peatlands has
been affected in other countries. For example, Minkkinen
et al. (2002) determined that 60% of Finnish peatlands
have been drained for forestry with 70 000–80 000 ha
undergoing drainage maintenance yearly. This has the
potential to increase export of DOC by 1Ð5 ð 109 g of
carbon per year.
Since the reduction in water table position in northern
peatlands under climate change is expected to result
from enhanced evapotranspiration rates, the water table
drawdown induced via drainage in this experiment is not
directly comparable to potential climate change scenarios.
However, since DOC concentrations in surface and pore
water were observed to increase in this study following
water table drawdown and remain elevated despite the
consistent loss of DOC in surface discharge, it is probable
that DOC concentrations in peatland waters will also
be increased under climate change. Moreover, changing
hydraulic gradients, hydraulic conductivity and patterns
of water table fluctuation resulting from peat subsidence
would occur regardless of the mechanism of water table
drawdown. The shift in the vegetation community along
the microtopographic gradient also resulted from the
drier site conditions. Thus, insight into DOC dynamics
under a climate change scenario can be gained through
comparison of E and D to C.
This study suggests that, if peatlands become drier,
enhanced water table fluctuations and vegetation productivity will lead to higher DOC concentrations in surface and pore water, particularly at wet microforms. An
increase in aerobic decomposition at dry microforms,
such as hummocks, may lead to reductions in DOC
pore water concentration in these zones. Thus, the initial hydrologic conditions and pattern of microtopography
may influence the response of peatland DOC dynamics
to climate change.
Higher DOC concentration in peatlands may help to
maintain DOC export from these ecosystems, despite
potential reductions in discharge resulting from higher
evapotranspiration rates (Moore, 1998), or lead to
increases in DOC export, as has been observed in Europe
(Freeman et al., 2001; Worrall et al., 2003) and predicted
in eastern Canada (Clair et al., 2002). Whereas other
studies have raised concerns that observed increases in
DOC export indicate a destabilization of peatland carbon stocks, this study suggests that high DOC export
may result from higher ecosystem productivity, potentially having little effect on net peatland carbon storage.
An increase in export of DOC to downstream ecosystems will increase acidity and reduce light penetration
depth (e.g. Schindler et al., 1996). DOC can also act as
a biogeochemical substrate (e.g. Wetzel, 1992); however,
if climate change results in a shift in DOC quality, then
its ability to act as substrate within the peatland and in
downstream ecosystems may be altered (Glatzel et al.,
2003). Further investigation is needed to determine the
effect of disturbance on peatland DOC quality.
ACKNOWLEDGEMENTS
We thank Karen Edmunston, Frank Dunnett, Erik Kellner, Matt Falcone, Leah Hartwig, Karola Tóth, Jamee
DeSimone, and J. R. van Haarlem for their assistance
in the field and laboratory. We also thank Nirom Peat
Moss for site access. Comments from T. Moore, M.
Waiser and an anonymous reviewer greatly improved
this manuscript. This research was funded by NSERC
(Canada) and Canadian Foundation for Climate and
Atmospheric Sciences (CFCAS) grants to JMW, NSERC
Julie Payette and CGS scholarships to MS and by the
federal Inter-Departmental Panel on Energy Research and
Development (PERD).
REFERENCES
Asada T, Warner BG. 2005. Surface peat mass and carbon balance in a
hypermaritime peatland. Soil Science Society of America Journal 69:
549– 562.
Baird AJ, Gaffney SW. 1995. A partial explanation of the dependency
of hydraulic conductivity on positive pore water pressure in peat soils.
Earth Surface Processes and Landforms 20: 561–566.
Belyea LR, Clymo RS. 2001. Feedback control of the rate of peat
formation. Proceedings of the Royal Society of London, Series B:
Biological Sciences 268: 1315– 1321.
Billett MF, Palmer SM, Hope D, Deacon C, Storeton-West R, Hargreaves KJ, Flechard C, Fowler D. 2004. Link land–atmosphere–
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
M. STRACK ET AL.
stream carbon fluxes in a lowland peatland system. Global Biogeochemical Cycles 18: GB1024. DOI: 10Ð1029/2003GB002058.
Blodau C, Moore TR. 2003. Experimental response of peatland carbon
dynamics to a water table fluctuation. Aquatic Sciences 65: 47–62.
Boelter DH. 1972. Water table drawdown around an open ditch in organic
soils. Journal of Hydrology 15: 329– 340.
Bourbonniere RA. 1989. Distribution patterns of dissolved organic matter
fractions in natural waters from eastern Canada. Organic Geochemistry
14: 97–107.
Brooks PD, McKnight DM, Bencala KE. 1999. The relationship between
soil heterotrophic activity, soil dissolved organic carbon (DOC)
leachate and catchment-scale DOC export in headwater catchments.
Water Resources Research 35: 1895– 1902.
Bubier JL, Frolking S, Crill PM, Linder E. 1999. Net ecosystem
productivity and its uncertainty in a diverse boreal peatland. Journal
of Geophysical Research—Atmospheres 104: 27683– 27692.
Christ MJ, David MB. 1996. Dynamics of extractable organic carbon in
spodosol forest floors. Soil Biology and Biochemistry 28: 1171– 1179.
Clair TA, Arp P, Moore TR, Dalva M, Meng FR. 2002. Gaseous carbon
dioxide and methane, as well as dissolved organic carbon losses from
a small temperate wetland under a changing climate. Environmental
Pollution 116: S143–S148.
Cleary J, Roulet NT, Moore TR. 2005. Greenhouse gas emissions from
Canadian peat extraction, 1990– 2000: a life cycle analysis. Ambio 34:
456– 461.
Clymo RS. 1984. The limits to peat bog growth. Philosophical
Transactions of the Royal Society of London, Series B: Biological
Sciences 303: 605– 654.
Devito KJ, Dillon PJ, Lazerte BD. 1989. Phosphorus and nitrogen
retention in five Precambrian shield wetlands. Biogeochemistry 8:
185– 204.
Foster DR, Wright HE, Thelaus M, King GA. 1988. Bog development
and landform dynamics in central Sweden and south-eastern Labrador,
Canada. Journal of Ecology 76: 1164– 1185.
Fraser CJD, Roulet NT, Moore TR. 2001. Hydrology and dissolved
organic carbon biogeochemistry in an ombrotrophic bog. Hydrological
Processes 15: 3151– 3166.
Freeman C, Evans CD, Monteith DT, Reynolds B, Fenner N. 2001.
Export of organic carbon from peat soils. Nature 412: 785.
Freeman C, Fenner N, Ostle NJ, Kang H, Dowrick DJ, Reynolds B,
Lock MA, Sleep D, Hughes S, Hudson J. 2004. Export of dissolved
organic carbon from peatlands under elevated carbon dioxide levels.
Nature 430: 195–198.
Freeze RA, Cherry JA. 1979. Groundwater. Prentice-Hall: Englewood
Cliffs, NJ.
Glatzel S, Kalbitz K, Dalva M, Moore T. 2003. Dissolved organic matter
properties and their relationship to carbon dioxide efflux from restored
peat bogs. Geoderma 113: 397– 411.
Gorham E. 1991. Northern peatlands: role in the carbon cycle and
probable responses to climatic warming. Ecological Applications 1:
182– 195.
Hillman GR. 1992. Some hydrological effects of peatland drainage
in Alberta boreal forest. Canadian Journal of Forest Research 22:
1588– 1596.
Hinton MJ, Schiff SL, English MC. 1997. The significance of storms for
the concentration and export of dissolved organic carbon from two
Precambrian Shield catchments. Biogeochemistry 36: 67–88.
Joiner DW, Lafleur PM, McCaughey JH, Bartlett PA. 1999. Interannual
variability in carbon dioxide exchanges at a boreal wetland
in the BOREAS northern study area. Journal of Geophysical
Research—Atmospheres 104: 27663– 27672.
Joosten H, Clarke D. 2002. Wise use of mires and peatlands—background
and principles including a framework for decision-making. International Mire Conservation Group and International Peat Society: Saarijarvi, Finland.
Kalbitz K, Popp P, Geyer W, Hanschmann G. 1997. ˇ-HCH mobilization
in polluted wetland soils as influenced by dissolved organic matter.
Science of the Total Environment 204: 37–48.
Kalbitz K, Solinger S, Park J-H, Michalzik B, Matzner E. 2000. Controls
on the dynamics of dissolved organic matter in soils: a review. Soil
Science 165: 277– 304.
Kang H, Freeman C, Ashendon TW. 2001. Effect of elevated CO2 on fen
peat biogeochemistry. Science of the Total Environment 279: 45–50.
Kellner E, Price JS, Waddington JM. 2004. Pressure variation in peat as a
result of gas bubble dynamics. Hydrological Processes 18: 2599– 2605.
Koprivnjak J-F, Moore TR. 1992. Sources, sinks, and fluxes of dissolved
organic carbon in subarctic fen catchments. Arctic and Alpine Research
24: 204–210.
Lu Y, Wassman R, Neue H-U, Huang C. 2000. Dynamics of dissolved
organic carbon and methane emissions in a flooded rice soil. Soil
Science Society of America Journal 64: 2011– 2017.
Lundquist EJ, Jackson LE, Scow KM. 1999. Wet–dry cycles affect
dissolved organic carbon in two California agricultural soils. Soil
Biology and Biochemistry 31: 1031– 1038.
Mattsson T, Kortelainen P, David MB. 1998. Dissolved organic carbon
fractions in Finnish and Maine (USA) lakes. Environment International
24: 521–525.
Mattsson T, Kortelainen P, Raike A. 2005. Export of DOM from boreal
catchments: impacts of land use cover and climate. Biogeochemistry
76: 373–394.
Minkkinen K, Korhonen R, Savolainen I, Laine J. 2002. Carbon balance
and radiative forcing of Finnish peatlands 1900– 2100—the impact of
forestry drainage. Global Change Biology 8: 785–799.
Moore PD. 2002. The future of cool temperate bogs. Environmental
Conservation 29: 3–20.
Moore TR. 1987. Patterns of dissolved organic matter in subarctic
peatlands. Earth Surface Processes and Landforms 12: 387– 397.
Moore TR. 1989. Plant production, decomposition, and carbon efflux in
a subarctic patterned fen. Arctic and Alpine Research 21: 156– 162.
Moore TR. 1998. Dissolved organic carbon: sources, sinks, and fluxes
and role in the soil carbon cycle. In Soil Processes and the Carbon
Cycle, Lal R, Kimble JM, Follett RF, Stewart BA (eds). Advances in
Soil Science. CRC Press: Boca Raton, FL; 281–292.
Moore TR, Dalva M. 2001. Some controls on the release of dissolved
organic carbon by plant tissues and soils. Soil Science 166: 38–47.
Moore TR, Roulet NT. 1993. Methane flux: water table relations in
northern wetlands. Geophysical Research Letters 20: 587–590.
Moore TR, Roulet NT, Waddington JM. 1998. Uncertainty in predicting
the effect of climatic change on the carbon cycling of Canadian
peatlands. Climatic Change 40: 229–245.
Mulholland PJ, Dahm CN, David MB, DiToro DM, Fisher TR, KögelKnabner I, Meybeck MH, Meyer JL, Sedell JR. 1990. What are the
temporal and spatial variations of organic acids at the ecosystem level?
In Organic Acids in Aquatic Ecosystems, Perdue EM, Gjessing ET
(eds). Life Sciences Research Report 48. Wiley: Chichester, UK;
315–329.
Pastor J, Solin J, Bridgham SD, Updegraff K, Harth C, Weishampel P,
Dewey B. 2003. Global warming and the export of dissolved organic
carbon from boreal peatlands. Oikos 100: 380–386.
Price JS. 2003. Role and character of seasonal peat soil deformation on
the hydrology of undisturbed and cutover peatlands. Water Resources
Research 39: 1241.
Price JS, Schlotzhauer SM. 1999. Importance of shrinkage and
compression in determining water storage changes in peat: the case
of a mined peatland. Hydrological Processes 13: 2591– 2601.
Qualls RG, Haines BL. 1991. Geochemistry of dissolved organic
nutrients in water percolating through a forest ecosystem. Soil Science
Society of America Journal 55: 1112– 1123.
Quinton WL, Roulet NT. 1998. Spring and summer runoff hydrology in a
subarctic patterned wetland. Arctic and Alpine Research 30: 285–294.
Roulet N, Moore T, Bubier J, Lafleur P. 1992. Northern fens: methane
flux and climatic change. Tellus Series B: Chemical and Physical
Meteorology 44: 100– 105.
Schiff S, Aravena R, Mewhinney E, Elgood R, Warner B, Dillon P,
Trumbore S. 1998. Precambrian shield wetlands: hydrologic control of
the sources and export of dissolved organic matter. Climatic Change
40: 167–188.
Schindler DW, Curtis JP, Parker BR, Stainton MP. 1996. Consequences
of climate warming and lake acidification for UV-B penetration in
North American boreal lakes. Nature 379: 705–708.
Schothorst CJ. 1977. Subsidence of low moor peat soil in the western
Netherlands. Geoderma 17: 265– 291.
Shurpali NJ, Verma SB, Kim J. 1995. Carbon dioxide exchange in a
peatland ecosystem. Journal of Geophysical Research—Atmospheres
100: 14319– 14326.
St-Arnaud C. 2007. Dynamique de la végétation d’un fen pauvre face à
une simulation de réchauffement climatique: réponses potentielles des
tourbières boréales à sphaignes. MSc thesis, Université Laval.
Steinberg CEW. 2003. Ecology of Humic Substance in Freshwaters.
Springer: New York.
Strack M, Waddington JM. 2007. Response of peatland carbon dioxide
and methane dynamics to a water table drawdown experiment. Global
Biogeochemical Cycles 21: GB1007. DOI: 10Ð1029/2006GB002715.
Strack M, Waddington JM, Rochefort L, Tuittila E-S. 2006. Response
of vegetation and net ecosystem carbon dioxide exchange at
different peatland microforms following water table drawdown.
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp
WATER TABLE DRAWDOWN AND PEATLAND DOC DYNAMICS
Journal of Geophysical Research—Biogeosciences 111: G02006. DOI:
10Ð1029/2005JG000145.
Thurman EM. 1985. Organic Geochemistry of Natural Waters. Martinus
Nijhoff/Dr. W. Junk: Dordrecht.
Tóth K. 2002. Dissolved organic carbon dynamics in a cutover and
restored peatland . MSc thesis, McMaster University.
Van Seters TE, Price JS. 2002. Towards a conceptual model of
hydrological change on an abandoned cutover bog, Quebec.
Hydrological Processes 16: 1965– 1981.
Waddington JM, Roulet NT. 1996. Atmosphere– wetland carbon
exchanges: scale dependency on CO2 and CH4 exchange on the
developmental topography of a peatland. Global Biogeochemical
Cycles 10: 233–245.
Waddington JM, Roulet NT. 1997. Groundwater flow and dissolved
carbon movement in a boreal peatland. Journal of Hydrology 191:
122– 138.
Waddington JM, Roulet NT. 2000. Carbon balance of a boreal patterned
peatland. Global Change Biology 6: 87–97.
Wetzel RG. 1992. Gradient-dominated ecosystems—sources and regulatory functions of dissolved organic matter in freshwater ecosystems.
Hydrobiologia 229: 181–198.
Whittington PN. 2005. The effects of water table draw-down on the
hydrology of a patterned fen peatland near Quebec City, Quebec,
Canada. MSc thesis, University of Waterloo.
Whittington PN, Price JS. 2006. The effect of water table draw-down
(as a surrogate for climate change) on the hydrology of a patterned
fen peatland near Quebec City, Quebec. Hydrological Processes 20:
3589– 3600.
Worrall F, Burt T, Shedden R. 2003. Long term records of riverine
dissolved organic matter. Biogeochemistry 64: 165– 178.
Copyright  2008 John Wiley & Sons, Ltd and Her Majesty the Queen in right of Canada.
Hydrol. Process. (2008)
DOI: 10.1002/hyp